© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 31-GEMMASCOPE-RES-16K
    11. 11961
    Prev
    Next
    INDEX
    Explanations

    threats and actions related to potential harm or danger

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    threats or statements of harm directed at someone or something.

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     hampir
    -0.49
     Numerade
    -0.46
     quase
    -0.45
     schein
    -0.45
     Almost
    -0.44
     nahezu
    -0.44
     nerede
    -0.43
     almost
    -0.43
    幾乎
    -0.43
     CanadaChoose
    -0.42
    POSITIVE LOGITS
     suicide
    0.75
     collapse
    0.70
     ruin
    0.68
     disaster
    0.66
     catastrophe
    0.64
     ruine
    0.63
    suicide
    0.62
     kill
    0.61
     annihilation
    0.60
     repris
    0.59
    Activations Density 0.564%

    No Known Activations